| Commit message (Collapse) | Author | Age | Files | Lines |
|---|
| ... | |
| | |
|
| | |
|
| | |
|
| |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
I've been doing a lot of work on mkinitrd's dmraid support lately some of the
lessons learned there also apply to the activation of (not used for /) dmraid
arrays in rc.sysinit, this will esp. help the livecd proerly activating and
using dmraid sets.
The patch fixes the following issues:
1) rc.sysinit (and mkinitrd used to) calls dm_resolve_name() on the output of
dmraid, this is not necessary as dm_resolve_name is only needed when the name
is /dev/dm-#, it is actually harmful as dm_resolve_name sometimes fails when
used like this, causing us to fail to activate dmraid sets
2) rc.sysinit does not remove the partitions from the raw disks used in a
raidset the initrd used to use special nash functionality to tell the kernels
to forget about the partitions of the raw disks it had scanned at boot. The
issue here is, say we have a mirror of sda and sdb, and that mirror has 2
partitions. Then the kernel when scanning the raw disks will find
sda1,sda2,sdb1,sdb2. But we do not want anything using those, they should use
the devicemapper versions as those are mirrored! I've even seen mounting by
LABEL / UUID do the wrong thing by using the raw partitions instead of the
raidset partitions.
I've recently done a patch to dmraid adding a --rm_partitions cmdline option
which makes dmraid do the ioctl which makes the kernel forget about the
partitions of the rawdisk (which in turn makes udev remove the device nodes).
This way we no longer need the nash magic.
Part of my rc.sysinit patch is adding --rm_partitions to the dmraid invocation
3) rc.sysinit uses dmraid -p, which tells dmraid to not setup device maps for
the partitions of the raidset and then kalls kpartx to setup the device maps
for the partitions. This can be simplified by removing the -p argument to
dmraid
4) with all these changes in place there is no longer a reason to loop, a
single dmraid invocation is enough.
p.s.
AFAIK with this patch the last user of dm_resolve_name is gone, so that
function can be removed. Better check twice though.
|
| | |
|
| |
|
|
| |
Noticed when reviewing (#448919)
|
| |
|
|
| |
bash 4 breaks the prior code.
|
| |
|
|
| |
(#481794, <sprabhu@redhat.com>)
|
| | |
|
| | |
|
| | |
|
| | |
|
| | |
|
| | |
|
| |
|
|
| |
No more magic files needed. In fact, I'm not sure they ever were. Whoops.
|
| |
|
|
|
|
|
| |
We were only reading the first line of pid files. Given that even
those first lines could have multiple pids, this *shouldn't* break
anything, but there may be some scripts with multi-line pid files
that aren't expecting this.
|
| |
|
|
| |
We'll see what breaks.
|
| |
|
|
|
|
| |
Notably, it prioritizes RAID/DM/LVM, etc. over the underlying block
devices. Given that initial RAID0 members share a UUID with the RAID
device itself, we need that.
|
| |
|
|
|
| |
Doing so blows away any label or UUID on the partition, which makes
specifying encrypted swap by label or UUID fail miserably.
|
| |
|
|
|
|
|
| |
<wwoods@redhat.com>)
Also, do some manipulation of the device name to (hopefully) handle things
like /dev/cciss/c0d0p1, and other devices with nested /dev hierarchies.
|
| |
|
|
|
| |
Just remove the compatibility case, and conflict with older alsa-utils
that would require it.
|
| |
|
|
| |
Avoids spurious 'rcX main process terminated with status 1' messages from upstart.
|
| |
|
|
| |
Checking whether or not it's a block device already covers that.
|
| |
|
|
| |
It's possible the initrd set the device up under a different LUKS name.
|
| | |
|
| | |
|
| | |
|
| |
|
|
| |
Also, fix permissions on rc.d/init.d/functions.
|
| |
|
|
| |
This reverts commit 2c55e1c2d7f4ee82aa0af9bfebfee889236c56bf.
|
| |\ |
|
| | |
| |
| |
| | |
cryptsetup should do this, by all rights. But it does not.
|
| |/
|
|
|
|
|
| |
We do need to explicitly load dm-mod for older kernels, because the
various tools won't necessarily do it. *sigh*
Reverts a chunk of 2285e2f27e1f9dce216a8b8791bd4f4237bdff80.
|
| |
|
|
| |
This matches the behavior of killproc.
|
| |
|
|
| |
This fixes the fact that the semantics of --details changed.
|
| | |
|
| |
|
|
| |
We need to pass different options depending on whether gdm will run or not.
|
| |
|
|
|
|
| |
If netfs isn't running from a tty (say, from NetworkManager), the attempt
to start a shell will fail, and we'll immediately fall through to rebooting.
That isn't good.
|
| |
|
|
|
| |
1) Loading dm-mirror, or dm-crypt, or whatever, will load it anyway.
2) We have udev. It creates device nodes. Why are we doing it by hand?
|
| | |
|
| |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
Changes to init scripts to better support stateless:
--rc.sysinit--
- don't update mtab in a r/o root
- mount on sys and /dev/pts returns EBUSY on a readonly root
- [ -n $CLIENTSTATE ] is always returning true, causing an unnecessary timeout
--network--
- /etc/mtab may not have any mount information in the readonly-root
cause, but the root mount is likely to be a network mount. Without
this, nfs hangs the reboot process.
--halt--
- don't try and update a read-only mtab
--rwtab--
- /var/empty/sshd/etc/localtime needs to be overwritten by sshd at
startup
- /var/lib/random-seed is written to at shutdown; this fixes the
error but doesn't return the functionality lost in a r/o /var/lib.
|
| |
|
|
|
|
|
|
| |
All networking scripts from /etc/sysconfig/network-scripts call logger without
full path. This can prevent system from booting if let's say dcc is installed
when compiled with wrong parameters and installs /usr/sbin/logger (see bellow)
It's not really a bug but this can be easily prevented. All /etc/init.d scripts
use full path for logger network-scripts should use it also
|
| | |
|
| |
|
|
| |
Updated iproute doesn't like the old syntax.
|
| | |
|
| | |
|
| | |
|
| |
|
|
| |
Also, remove some old rhgb code.
|
| | |
|
| | |
|
| | |
|
